BEDROCK STRENGTH CONTROLS ON THE VALLEY MORPHOMETRY OF BIG CREEK, IDAHO AND VALLEY COUNTIES, CENTRAL IDAHO

The fluvial geomorphology of Big Creek, a tributary to the Middle Fork Salmon River in central Idaho, was investigated using the Schmidt Hammer to determine what, if any, relationships exist between the relative in situ bedrock strength and valley morphometry, and through which mechanisms or processes they might be related. The Schmidt Hammer is a useful tool in the quantitative study of geomorphology since it measures rock strength by quantifying relative in situ rebound values of exposed bedrock. Thirteen reaches, ranging from 200 m to 1700 m in length, are studied on Big Creek. Data were measured in the field and extracted from digital elevation models (DEMs) in order to compare rock strength to three valley parameters: valley floor width, stream gradient, and hillslope gradient. Valley floor width was measured with a laser rangefinder. Stream gradient and hillslope gradient were measured from DEMs using ArcMap, SWAT, and RiverTools software.

A strong negative correlation (r² = 0.8394, P = 1.08x10-5) exists between Schmidt hammer rebound and valley floor width; wide valley floors correspond to weak bedrock, and narrow valley floors correspond to strong bedrock. The valley floor width is most responsive to the rebound value on the weaker of the two valley sides (usually the north side). A statistically significant difference in rebound values exists between north- and south-facing slopes, indicating that aspect has an influence on bedrock strength. Rebound values on the north side of the valley, which is south-facing, tend to be lower than those on the south side of the valley, caused by the north side of the valley experiencing more freeze-thaw cycles.

No correlation exists between Schmidt Hammer rebound and stream gradient, possibly because of the masking effect that the thin veneer of alluvial sediment has on the gradient of underlying bedrock. Likewise, no correlation exists between Schmidt Hammer rebound and hillslope gradient, which is probably due to the complexity of hillslope processes and the limited spatial distribution of rebound data. However, hillslope gradient is weakly correlated (r² = 0.6473, P = 0.0009) to valley floor width.